SA Schools data exploration

Overview

Data exploration using the data prepared by combining the “schools master lists” from the Department of Basic Education website for schools in South Africa.

# LIBRARIES ----------------------------------------------------------------------------------------
library(dplyr)

Attaching package: ‘dplyr’

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

library(here)

here() starts at /home/megan/megan/sa_schools

library(ggplot2)
library(plotly)

Attaching package: ‘plotly’

The following object is masked from ‘package:ggplot2’:

    last_plot

The following object is masked from ‘package:stats’:

    filter

The following object is masked from ‘package:graphics’:

    layout

# READ DATA ----------------------------------------------------------------------------------------
sa_schools <- readRDS(here::here("data/03_sa_schools.RDS"))

Learner and teacher numbers

# VISUALISE ----------------------------------------------------------------------------------------
p <- ggplot(sa_schools, aes(x = Learners)) +
  geom_histogram() +
  labs(title='Histogram of no. of learners',
       x= 'No. of learners', y = 'School count')
ggplotly(p)

`stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Removed 309 rows containing non-finite values (stat_bin).

p <- ggplot(sa_schools, aes(x = Educators)) +
  geom_histogram() +
  labs(title='Histogram of no. of educators',
       x= 'No. of educators', y = 'School count')
ggplotly(p)

`stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Removed 683 rows containing non-finite values (stat_bin).

Country and provincial characteristics

sa_schools <- sa_schools %>%
  mutate(Quintile = factor(sa_schools$Quintile))

Quintiles

p <- ggplot(sa_schools, aes(x = Quintile, fill = Quintile)) +
  geom_bar() +
  labs(title='Distribution of schools in South Africa across "quintiles"',
       x= 'Quintile', y = 'School count') +
   scale_fill_brewer(palette="Set1", direction = 1, na.value = "grey")
ggplotly(p)

sa_schools_quint <- sa_schools %>%
  group_by(Province, Quintile) %>%
  summarise(count = n()) %>%
  #filter(Quintile != "NA") %>%
  mutate(Quintile = factor(Quintile, levels = c("5", "4", "3", "2", "1", "NA")))

Factor `Quintile` contains implicit NA, consider using `forcats::fct_explicit_na`

p <- ggplot(sa_schools_quint) +
  geom_bar(aes(x = Province, y = count, fill = Quintile), stat="identity") +
  scale_fill_brewer(palette="Set1", direction = -1, na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1)) +
  labs(title='Distribution of schools in each province in South Africa across "quintiles"',
       x= '', y = 'School count')
ggplotly(p)

Rural vs Urban

p <- ggplot(sa_schools, aes(x = Urban_Rural, fill = Urban_Rural)) +
  geom_bar() +
  labs(title='Rural vs urban schools in South Africa',
       x= '', y = 'School count') +
   scale_fill_brewer(palette="Set1", direction = 1, na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1))
ggplotly(p)

sa_schools_ur <- sa_schools %>%
  group_by(Province, Urban_Rural) %>%
  summarise(count = n()) 

Factor `Urban_Rural` contains implicit NA, consider using `forcats::fct_explicit_na`

p <- ggplot(sa_schools_ur) +
  geom_bar(aes(x = Province, y = count, fill = Urban_Rural), stat="identity") +
  scale_fill_brewer(palette="Set1", direction = -1,  na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1)) +
  labs(title='Urban vs Rural schools in each province in South Africa"',
       x= '', y = 'School count')
ggplotly(p)

Sector type

p <- ggplot(sa_schools, aes(x = Sector, fill = Sector)) +
  geom_bar() +
  labs(title='Sector type of schools in South Africa',
       x= '', y = 'School count') +
   scale_fill_brewer(palette="Set1", direction = 1, na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1))
ggplotly(p)

sa_schools_sec <- sa_schools %>%
  group_by(Province, Sector) %>%
  summarise(count = n()) 
p <- ggplot(sa_schools_sec) +
  geom_bar(aes(x = Province, y = count, fill = Sector), stat="identity") +
  scale_fill_brewer(palette="Set1", direction = -1,  na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1)) +
  labs(title='Sector type schools in each province in South Africa"',
       x= '', y = 'School count')
ggplotly(p)

Phase

p <- ggplot(sa_schools, aes(x = Phase, fill = Phase)) +
  geom_bar() +
  labs(title='Distribution of phase of schools in South Africa',
       x= '', y = 'School count') +
   scale_fill_brewer(palette="Set1", direction = 1, na.value = "grey")  +
  theme(axis.text.x = element_text(angle = 40, hjust = 1))
ggplotly(p)

sa_schools_phase <- sa_schools %>%
  group_by(Province, Phase) %>%
  summarise(count = n()) 
p <- ggplot(sa_schools_phase) +
  geom_bar(aes(x = Province, y = count, fill = Phase), stat="identity") +
  scale_fill_brewer(palette="Set1", direction = -1,  na.value = "grey") +
  theme(axis.text.x = element_text(angle = 40, hjust = 1)) +
  labs(title='Split between phase of schools in each province in South Africa"',
       x= '', y = 'School count')
ggplotly(p)